This two-year research proposal continues our attempts for over a decade to reduce alcohol abuse among university students. We have tested a variety of theory-driven interventions, including a) educating students about their overestimation of peer drinking norms, b) teaching server intervention to the hosts of fraternity parties, c) providing BAC feedback in party settings, d) manning booths at fraternity parties to illustrate negative consequences of excessive alcohol consumption, and e) teaching partygoers simple performance games they can use on themselves and others to estimate whether they are too impaired to drive. Although some of these interventions influenced a significant reduction in self-reported alcohol consumption and concomitant negative outcomes, none influenced a significant reduction in actual blood alcohol concentration (BAC) at parties. The proposed research takes a completely different and straightforward approach to reducing excessive alcohol consumption at fraternity parties. Our pilot testing of this approach, based on theories of reinforcement and controlled drinking, showed remarkable effects. Specifically, the incentive/reward technique we propose to test comprehensively reduced mean BAC at two fraternity parties significantly to .063 (n=86) and .056 (n=91), respectively from baseline means of .092 (n=96) and .102 (n=83) at two prior parties held by the same fraternity. Moreover, the incentive/reward intervention reduced the number of partygoers who were legally drunk (BAC equal to or greater than .08) from 52 (54.2 percent) and 52(62.7 percent) at the two baseline parties to 26 (30.2 percent) and 20 (22 percent) at the two intervention parties. The incentive/reward intervention simply enabled partygoers to enter a $100 cash raffle if his or her BAC, measured between the hours of 11:30pm and 12:30am, was below .05. Pretest-posttest group randomized trial methodology (Murray, 1998) is proposed to demonstrate that a particular intervention package will be effective across a number of randomly selected fraternities. The proposed research also evaluates whether repeated exposures to parties with an incentive/reward contingency to hold BAC below .05 can change social norms with regard to party drinking and lead to long-term maintenance of lowered intoxication at fraternity parties.